Download as pdf or txt
Download as pdf or txt
You are on page 1of 9

VISVESVARAYA TECHNOLOGICAL UNIVERSITY,

BELAGAVI - 590018

A Technical Seminar Report on


“PERSON DETECTION FOR SOCIAL DISTANCING AND
SAFETY VIOLATION ALERT BASED ON SEGEMTNED
ROI”

Submitted in partial fulfillment of the requirements as per VTU curriculum of


BACHELOR OF ENGINEERING
IN
COMPUTER SCIENCE & ENGINEERING
By

GUDUGUNTI ANIRUDH 4AL18CS023

Under the Guidance of


Mr. K Surendra
Senior Assistant Professor

DEPARTMENT OF COMPUTER SCIENCE & ENGINEERING


ALVA’S INSTITUTE OF ENGINEERING AND TECHNOLOGY
MOODBIDRI-574225, KARNATAKA
2021– 2022
ALVA’S INSTITUTE OF ENGINEERING AND TECHNOLOGY
MIJAR, MOODBIDRI D.K. -574225
KARNATAKA

DEPARTMENT OF COMPUTER SCIENCE & ENGINEERING

CERTIFICATE

This is to certify that GUDUGUNTI ANIRUDH (4AL18CS023) has submitted


Technical Seminar Report on “Person Detection for Social Distancing and Safety
Violation Alert based on Segmented ROI” for the VII semester B.E. in Computer
Science & Engineering during the year 2021-22

Mr. K Surendra Mr. K Surendra Dr. Manjunath Kotari

Seminar Guide Seminar Coordinator Professor and Head


ALVA’S INSTITUTE OF ENGINEERING AND TECHNOLOGY

MIJAR, MOODBIDRI D.K. -574225

KARNATAKA

DEPARTMENT OF COMPUTER SCIENCE & ENGINEERING

Declaration

I, GUDUGUNTI ANIRUDH hereby declare that the dissertation entitled Person


Detection for Social Distancing and Safety Violation Alert based on Segmented
ROI, is completed and written by us under the supervision of my guide Mr. K
Surendra, Assistant Professor, Department of Computer Science and Engineering.
Alva’s Institute of Engineering and Technology, Moodbidri, DEPARTMENT OF
COMPUTER SCIENCE & ENGINEERING of the VISVESVARAYA
TECHNOLOGICAL UNIVERSITY, BELAGAVI during the academic year 2021-
2022.The dissertation report is original and it has not been submitted for any other
degree in any university.

GUDUGUNTI ANIRUDH
4AL18CS023
ACKNOWLEDGEMENT

The satisfaction and euphoria that accompany a successful completion of any task would
be incomplete without the mention of people who made it possible, success is the epitome
of hard work and perseverance, but steadfast of all is encouraging guidance.

So, with gratitude I acknowledge all those whose guidance and encouragement
served as beacon of light and crowned the effort with success.

I thank my Seminar Guide Mr. K Surendra, Senior Assistant Professor in


Department of Computer Science & Engineering, who has been my source of inspiration.
He has been especially enthusiastic in giving his valuable guidance and critical reviews.

The selection of Technical Seminar Topic as well as the timely completion is mainly
due to the interest and persuasion of my Seminar Coordinator Mr. K Surendra, Senior
Assistant Professor, Department of Computer Science & Engineering. I will remember his
contribution for ever.

I sincerely thank, Dr. Manjunath Kotari, Professor and Head, Department of


Computer Science & Engineering who has been the constant driving force behind the
completion of the seminar.

I thank Principal Dr. Peter Fernandes, for his constant help and support
throughout.

I am also indebted to Management of Alva’s Institute of Engineering and


Technology, Mijar, Moodbidri for providing an environment which helped me in
completing the seminar.

Also, I thank all the teaching and non-teaching staff of Department of Computer
Science & Engineering for the help rendered.

Finally I would like to thank my parents and friends whose encouragement and
support was invaluable

GUDUGUNTI ANIRUDH
4AL18CS023

i
ABSTRACT
In tending to the worldwide Covid-19 pandemic situation, the process of flattening the
curve for coronavirus cases will be difficult if the citizens do not take action to prevent the
spread of the virus. One of the most important practices in these outbreaks is to ensure a
safe distance between people in public. This study focuses on detecting people in areas of
interest using the Mobile Net Single Shot Multibox Detector (SSD) object tracking model
and OpenCV library for image processing. The distance will be computed between the
persons detected in the captured footage and then compared to a fixed pixels’ values. The
distance is measured between the central points and the overlapping boundary between
persons in the segmented tracking area.
With the detection of unsafe distances between people, alerts or warnings can be issued to
keep the distance safe. In addition to social distance measure, another key feature of the
system is detecting the presence of people in restricted areas, which can also be used to
trigger warnings. Some analysis has been performed to test the effectiveness of the program
for both purposes. From the results obtained, the distance tracking system achieved
between 56.5% to 68% accuracy for testing performed on outdoor and challenging input
videos, while 100% accuracy was achieved for the controlled environment on indoor
testing.
TABLE OF CONTENTS

CHAPTER DESCRIPTION PAGE


NO. NO.
ACKNOWLEDGEMENT i

LIST OF FIGURES iii

LIST OF TABLES iv
1. INTRODUCTION 1
1.1 PROBLEM STATEMENT 2
1.2 OBJECTIVE 2

2. RELATED WORK 3-4

3. PROPOSED METHODOLOGY 5
3.1 OBJECT DETECTION MODEL 6
3.2 THREADING PARALLELISM 6
3.3 MASKING FRAME FOR ROI AREA ESTIMATION 6

3.4 DETERMINE PERSON LOCATION 7

3.5 CALCULATE THE CENTRE POINT OF 7

A BOUNDING BOX
3.6 CALCULATE DISTANCE BETWEEN BOUNDING 8

BOX

4. RESULT ANALYSIS 8
4.1 MASKED ROI 8
4.2 PERSON DETECTION LOCALIZATION AND SOCIAL 9-13
DISTANCING
5. CONCLUSION 14

6. REFERENCES 15
LIST OF FIGURES

FIG NO. FIGURE NAME PAGE


NO.

3 PERSON DETECTION FOR SOCIAL DISTANCING AND SAFETY 5


VIOLATION ALERT BASED ON SEGMENTED ROI FLOWCHART.

3.1 THE GROUND PLANE OF A HUMAN WALKING MODEL. 7

4.1 OPENCV MASKING STEP TO GET FOREGROUND FRAME. 9

4.2 MASKED BACKGROUND FRAME WITH COMBINATION OF 9


FOREGROUND FRAME

4.3 GREEN BOUNDING BOXES WHEN THERE IS NO SOCIAL DISTANCE 9


VIOLATION.
4.4 RED BOUNDING BOXES INDICATE AN ALERT OR WARNING WHEN 10
THERE ARE SOCIAL DISTANCE VIOLATIONS.

4.5 MULTIPLE RED BOUNDING BOXES FOR A SINGLE PERSON SHOW 11


THE OBJECT DETECTION FAILS TO LOCATE PERSON ACCURATELY
ON DATASET PETS2009
THE OBJECT DETECTION FRAMEWORK FAILS TO LOCATE THE
4.6 LOCATION OF THE PERSON IN THE FRAME ON DATASET TOWN 12
CENTRE.

4.7 RED BOUNDING BOXES WITH THE WORD “TRESPASS” INDICATE AN 12


ALERT OR WARNING WHEN ENTERING THE ROI.

4.8 NO RED BOUNDING BOXES ON DETECTED PERSONS OUTSIDE THE 13


ROI.

iii
LIST OF TABLES

Table Description Page number


no

2.1 MAP SCORE RESULT OF FRONTIER MODELS ON COCO 4


DATASET

4.1 ACCURACY OF SOCIAL DISTANCE MONITORING. 11

4.2 ACCURACY OF ALERT FOR RESTRICTED AREA BASED ON 13

ROI.

iv

You might also like